Here is the plot of two DAC's. The older Audigy 2NX vs a new MUSE (Hong-Kong auction special) . Newer DAC's are cleaner. Both 24 bit, 96K.
The MUSE has BurrBrown chips, digital and analog. No idea what is in the Audigy. I look forward to my Wolfson based auction DAC any day.
The MUSE has BurrBrown chips, digital and analog. No idea what is in the Audigy. I look forward to my Wolfson based auction DAC any day.
An externally hosted image should be here but it was not working when we last tested it.
If I'm reading the graph correctly, both DACs have noise components that are at least 90 db below the reference 1 kHz tone.
If thats the case then that level of noise for either DAC is inaudible. So even though there are differences in the noise levels between each DAC, both are at levels that would not be audible.
You are reading it correctly. However, the do sound different. (not so much through speakers, but Grado's yes) My goal in this exercise is to establish my test bench better than the equipment I am playing with. For speaker building, heck 60 dB was fine. I have started playing with tubes, DAC's, and op-amps again, and not need much better instrumentation. This is close to where I need to be for tube work.
Off topic as we don't get news from your side of the globe, in GM's rush to kill itself, is Holden still going to make it? Pontiac, Saturn, Saab, Opel dead.
Off topic as we don't get news from your side of the globe, in GM's rush to kill itself, is Holden still going to make it? Pontiac, Saturn, Saab, Opel dead.
Noise. Not sure. When comparing the MUSE to an older NAD and a new ROTEL CD player, it's overall tonal balance was very close, but the MUSE had much more detail, including detail in the recording noise and garbage. Both good and bad. The Audigy just seemed "limited" and yes, maybe nosier. One thing is it goes out to lunch at .8 V out RMS. I am using it for the A2D in both tests, so how much of the residual in both traces is the A2D, I don't know. Looking for another to compare it to. M-Audio FW 610 maybe. I really need a better mic preamp anyway ( Audio Buddy overloads easy and oscillates at 35K). I want a 2 channel mic preamp with line in to USB so I have gain controls for my SopundEasy probes as well as the mic. Recommendations greatly appreciated.
If there was a sound difference between DACs, then its either frequency response errors or audible noise. Not sure what else could be left as a contributing factor after those two.
Well, there's no practical listening environment that could render audible that noise floor. So, I'd say don't sweat it, either DAC is fine, and equivalent for use.
Holden Australia is probably considered by the government to be "too big to fail" in the sense that they are one of two remaining big car manufacturers (the other is Ford). I'm sure taxpayer money will be found if necessary to keep them afloat, if things turn really ugly.
Long standing debate on what to measure and why different pieces sound different so I won't dig into that here. But, in playing with different phase splitter tubes and topologies, I found the Audigy was masking differences in the third harmonic distortion causing difficulty in my measurements on AC balance vs output bias balance caused distortion. Being an old lab rat, I still believe in the test set needing to be 10 times cleaner than what you are measuring. Measuring solid state is a diferent problem. Even the worst op-amp is better than any bench I can afford. I have been able to capture some off frequency compression of short pure pulse signals in power amps. I need more resolution to be sure of what I am seeing is real or a measurement artifact.
I can't speak on small equalization errors, as TrueRTA can calibrate in a loopback test, but that is applied to the measurement, not to real time for listening.
Glad to hear Holden may hold on. They seem to still put a tad bit of local personality in their cars. I miss that in "world" cars. I finally sold the last of my British cars so I can afford to play in audio again. Just too scary to drive them in Maryland roads.
I can't speak on small equalization errors, as TrueRTA can calibrate in a loopback test, but that is applied to the measurement, not to real time for listening.
Glad to hear Holden may hold on. They seem to still put a tad bit of local personality in their cars. I miss that in "world" cars. I finally sold the last of my British cars so I can afford to play in audio again. Just too scary to drive them in Maryland roads.
DAC chips sound different, and it's in timbre or tonal quality mainly. I don't think those two simplistic parameters can remotely suffice to explain this 🙂
But I am sure Andrea, your evaluations of op-amps have some to do with it!
Sure surprised me how good they have gotten. Detail, resolution and dynamics. Now, to find something we can measure to explain them.
But for this, I am only worried about measurements. Looks like the e-mu 1616 is what I want. Just have to find a way to justify $500 for my bench.
What I did not post is that the Audigy left channel input has 10 times the distortion as the right.
Sorry about the off topic, but in Washington DC, if it is news outside 20 miles, it is not in the paper. Gad how I love this place. 😀
I probably should have mentioned THD measurements as well.
But as to timbre or tonal quality, thats just frequency response changes.
No need to look for ghosts, the correlation is perfectly clear and understandable between those measures and audible differences.
Audible differences between two well designed DACs with the same bit length (at lest 16 bits) and sample frequency (at least 44,100) is wishful thinking.
Nothing like that, sorry. It's the same with (all) opamps... would you say they have funny frequency responses? 🙂
Last edited:
Well, if you are talking about tonality – then that is frequency response changes that deviate from a flat response. After all, that’s what tone controls do, vary the frequency response.
So, if audible tone variations (what you call tonality) do not come from frequency response variations, then where do they come from?
I hope to find out soon. It would be great if you are correct and the parts are now all that good and we can quit worrying about it and pay attention to important things like ground impedance. Op-Amps are almost there. I ordered a Wolfson based head amp/DAC to compare to my Muse BurrBrown and Behringer AK Dac's. The MUSE has the same op-amps and one version newer DAC as my Rotel 1070 and I can't tell them apart easily. Just a tad in background noise that I suspect can be attributed to the MUSE cheap execution of the power supply, and I can only hear it through the Grado's, not my speakers. Next to my NAD 530, you could hear the difference on cheap speakers from across the room. There is something else, a level of detail that is not pumped up midrange or level matching. Same tonal balance, but a difference in resolution. I measured some. All are flatter than my measurement system can do. All these new DAC's are better than my A2D. So looking for a newer A2D. E-mu looks promising.
I call "tonality" the colors in sound, literally. Because I hear colors in sound; independently from the basic "dark" or "bright" character of it.
DAC chips & opamps sound equally different from each other in their reproduction of sonic colors.
Where do they come from? Harmonics... (i.e. harmonic content)
Last edited:
- Status
- Not open for further replies.